I have noticed some slight bouncing on the Coolant temp and oil pressure gauges on my MS1’ed 91 with a recently installed fresh 96 1.8 motor. It turned out to be some old connectors on the sensors that I replaced which fixed the gauge issue. As part of my troubleshooting I looked at the voltages recorded in the MS logs and noticed a swing in the voltage of around .4-.5 volts, up to .8 volts at higher rpms. I have looked over all my grounds, cleaned connectors, added ground wires to block and ran an additional 2 gauge ground wire directly from the battery to ECU, wideband and engine/starter/alt. No changes were noticed. I even swapped the 1.8 pulley onto the old 1.6 alternator to see if it was the alternator causing the problems. It still displayed the same change in voltage output, so I am not too sure what is going on. I was planning on replacing the alternator since I have pretty much replaced everything else as part of installing the 1.8 engine, but now I am wondering if something else is causing the problem. Or is it even a problem. I am just guessing that a swing in voltage of 0.5V should be enough to impact the measurements taken by the wideband and maybe even some of the MS readings/responses.
Searches have not yielded any helpful information. I have seen some threads of people having larger voltage spikes due to bad or loose grounds, but that does not seem to be my problem.
I added a screenshot from my latest log that shows a voltage swing of up to 0.5V within a few time stamps at cruise rpm and up to 0.8V at >5500rpm. The car runs fine with smooth power delivery and cruise. I want to track my car again this summer and want to have all issues worked out before then.

 

NA alternators are crap. They usually output about 13.2V at the ECU, which drops to 12V at idle with both fans on etc. On bigturbo's 400hp 1.6, it drops to ~11.5V during WOT pulls.

 

The numbers Reverant quotes are typical to what I see. I cruise around 13.8v and with the a/c on it drops to something like 12.5v.

When my alternator was bad, when idling with a/c on, I could watch my voltage slowly drain off. 12.5,12.4,12.3...all the way to 11.5v and continuing to drop.


I've always thought about adding a 2-farad cap to my stereo install, since it causes my lights to dim anyways and hopefully as a side effect help stabilize my voltages with the a/c and in boost.

 

I am not too worried about the voltage drop as much as the fluctuation in voltage which appears as noise when you graph the log output. I am installing a noise filter used on stereo systems to clean up the signal to the MS and wideband. I just want to make sure I am not going to hurt any other components with that much noise on the alt output.

I never see the voltage drop below 13v so it is providing charge as needed, even with AC, fans, headlight and laptop inverter running.
I also noticed that all three of my inverters will give faults when coming to a stop (idling) after a longer dive (20-30min). It usually coincides with larger voltage spikes or droops, which adds to the confusion. Each inverter gives me a fault in the same situation. I fist thought it was the one old inverter, but with all of them acting up it has me wondering.
Are there other alternators that would be relatively easy to swap in? I don't mind doing some fab work to make it fit. I have no idea what other vehicles use the same belt pattern and have a 1 wire alt or something compatible with the miata system.
I recently developed a Airbag light flash (6 flashes for module failure or clockspring issues). I have no idea if it is related, but it did appear at the same time I started noticing these issues. I have a spare clock spring that I will be swapping in to see if it takes care of the Airbag light. Either that or I might finally have the excuse to install the removable wheel I have been wanting.

 

I'm a developing a board to make it possible to swap in a 99-05 alternator. That will keep the system voltage stable to 14.2, +/-0.1V.

 

Here is where I graphed a 3 second section from my last log.

http://carphotos.cardomain.com/ride_...4002_large.jpg

That appears like a lot of noise to me. I might just be over thinking this one, but that much of a swing has to impact electronic measurements that use reference voltages.

 

Are the NB alternators just externally regulated or does it need other control inputs from the ECU?

 

Regulated by the ECU.

 

Try gooogling "Lightweight Alternators", came across a few solutions here in the UK.

 

FWIW a Speed / Voltage log of a normally aspirated Miata during an autocross run. At first the car is at idle sitting at the starting line. GPS jitter shows a couple MPH but that is not the case. DL1 powered from the lighter socket.

https://www.miataturbo.net/attachmen...1&d=1300846442

 

sjmarcy, looking at the first 27 seconds of that chart your voltages only vary between 14.02 and 13.92V (0.1V swing) which is a pretty tight band. I am assuiming you were close to idle at that point. I could live with that. The voltages I show in the link above is only over a 3 second period (I changed the time stamps for scaling) and I am seeing 0.8V swings. I could deal with the change in voltage as you change engine RPM, my concern is that sitting at constant idle or even constant cruise RPM, my voltages bounce all over the place.
I might go junk yard hunting this weekend to see if I can find alternative alternators.

 

Do the rx-7 alternators that people use have better voltage outputs?

 

more amperage IIRC, voltage should always be ~14v.


Add smoothing to Sjmarcy's graph and it doesn't look bad at all.

 

RX7 alts are ever worse. They output 14V when cold, but drop as low as 12.8V if heatsoaked. They also tend to put more load on the engine at idle - my idle quality increased significantly once I swapped in the stock alternator.

 

Not necessarily. Since circuits powered by that source get filtered and cleaned up. For instance, various 5V reference signals derive from dedicated circuits designed to do just that, provide a very steady 5 volts. Despite a supply varying within expected amounts.

 

Here are the voltages seen during 3 autocross runs…they are smoothed and scaled. Seems acceptable to me. Normally aspirated car, stock ecu, etc. There is more load with high rev WOT action, as expected. You could run one of those car audio "stiffening" capacitors to smooth out the fluctuations if you so desired. The orange run had a mechanical issue mid course and so the car was stopped…note that the voltage returned to the idle style again.

https://www.miataturbo.net/attachmen...1&d=1300893440

 

You'll laugh when you see my voltage log in boost.

 

Here's my old Elise in ABS…those devices love to guzzle current!

I didn't get to try this out…but if more current were available to the ABS solenoids…their reactions should improve…and perhaps the braking effect. By avoiding such a strong sag…the ABS power would rise maybe 20-30%.

 

Ok, I will quit worrying then. I am adding the noise filter to my MS and Wideband power feeds though. Thanks for the help. I will report back once I get the filter installed.
Would it be possible to install a external voltage regulator on the alternator output?